Hugging Face model family · google
gemma-4-31B model family
Trace gemma-4-31B and its 1,214 community derivatives back to the base model. Model Dig maps the lineage of GGUF quantizations, merges, adapters and fine-tunes, ranked by download heat and likes — so you can see exactly how each variant relates to gemma-4-31B.
Open interactive lineage graph → Browse all families
1,215 models in family
576 GGUF variants
44,268 base downloads
2026-03-12 base released
Popular gemma-4-31B derivatives
| # | Model | Author | Downloads | Likes |
|---|---|---|---|---|
| 1 | gemma-4-31B-it | ⬇ 6,614,519 | ♥ 4,355 | |
| 2 | Gemma-4-31B-IT-NVFP4 NVFP4 | nvidia | ⬇ 1,140,518 | ♥ 588 |
| 3 | gemma-4-31B-it-AWQ AWQvLLM | QuantTrio | ⬇ 665,541 | ♥ 14 |
| 4 | gemma-4-31B-it-qat-NVFP4-Blackwell NVFP4QATvLLM | melcheikh | ⬇ 588,029 | ♥ 4 |
| 5 | gemma-4-31B-it-AWQ-4bit AWQ4-bit | cyankiwi | ⬇ 450,426 | ♥ 55 |
| 6 | gemma-4-31B-it-GGUF GGUF | unsloth | ⬇ 429,970 | ♥ 678 |
| 7 | gemma-4-31B-it-FP8-block FP8vLLM | RedHatAI | ⬇ 427,430 | ♥ 52 |
| 8 | gemma-4-31B-it-assistant | ⬇ 195,170 | ♥ 344 | |
| 9 | gemma-4-31B-it-qat-w4a16-ct QATW4A16 | ⬇ 144,226 | ♥ 102 | |
| 10 | gemma-4-31B-it-MLX-8bit 8-bitMLX | lmstudio-community | ⬇ 138,467 | ♥ 13 |
| 11 | gemma-4-31B-it-MLX-4bit 4-bitMLX | lmstudio-community | ⬇ 136,512 | ♥ 2 |
| 12 | gemma-4-31B-it-MLX-6bit 6-bitMLX | lmstudio-community | ⬇ 131,538 | ♥ 2 |
| 13 | gemma-4-31B-it-MLX-5bit 5-bitMLX | lmstudio-community | ⬇ 130,720 | ♥ 2 |
| 14 | gemma-4-31B-it-NVFP4 NVFP4vLLM | RedHatAI | ⬇ 124,289 | ♥ 60 |
| 15 | gemma-4-31B-it-qat-q4_0-gguf GGUFQ4_0QAT | ⬇ 115,066 | ♥ 152 | |
| 16 | gemma-4-31B-it-ARA-heretic-GGUF GGUFHereticARA | mradermacher | ⬇ 105,256 | ♥ 248 |
| 17 | gemma-4-31B-it-QAT-GGUF GGUFQAT | lmstudio-community | ⬇ 91,542 | ♥ 10 |
| 18 | Gemma4-31B-QAT-Uncensored-HauhauCS-Balanced-MTP MTPQATUncensored | HauhauCS | ⬇ 91,431 | ♥ 209 |
| 19 | gemma-4-31B-it-FP8-dynamic FP8DynamicvLLM | RedHatAI | ⬇ 85,708 | ♥ 25 |
| 20 | gemma-4-31B-it-qat-GGUF GGUFQAT | unsloth | ⬇ 80,094 | ♥ 226 |
| 21 | gemma-4-31B-it-uncensored-heretic-GGUF GGUFHereticUncensored | llmfan46 | ⬇ 72,543 | ♥ 209 |
| 22 | gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-GGUF GGUFHereticUncensored | llmfan46 | ⬇ 59,347 | ♥ 91 |
| 23 | gemma-4-31B-it-GGUF GGUF | ggml-org | ⬇ 53,717 | ♥ 48 |
| 24 | gemma-4-31B-it-speculator.eagle3 Speculator | RedHatAI | ⬇ 51,867 | ♥ 54 |
| 25 | google_gemma-4-31B-it-GGUF GGUF | bartowski | ⬇ 44,117 | ♥ 98 |
| 26 | gemma-4-31B-it-GGUF GGUF | lmstudio-community | ⬇ 43,269 | ♥ 31 |
| 27 | gemma-4-31B-it-heretic-i1-GGUF GGUFHeretici1 | mradermacher | ⬇ 36,246 | ♥ 59 |
| 28 | gemma-4-31B-it-int4-AutoRound AutoRoundINT4 | Intel | ⬇ 36,164 | ♥ 55 |
| 29 | Gemma-4-31B-JANG_4M-CRACK-GGUF GGUFCRACKJANG | douyamv | ⬇ 35,802 | ♥ 249 |
| 30 | gemma-4-31B-it-oQ6e-mtp MTPoQ6e | Jundot | ⬇ 23,729 | ♥ 3 |
| 31 | Gemma-4-Dark-Orthia-Untied-31B-GGUF GGUF | QuantaPlanta | ⬇ 22,467 | ♥ 8 |
| 32 | gemma-4-31B-it-unsloth-bnb-4bit 4-bitbitsandbytes | unsloth | ⬇ 21,495 | ♥ 24 |
| 33 | gemma-4-31B-it-The-DECKARD-HERETIC-UNCENSORED-Thinking-i1-GGUF GGUFHeretici1Uncensored | mradermacher | ⬇ 20,207 | ♥ 17 |
| 34 | Gemma-4-31B-Fable-Coder-GGUF CoderGGUF | mradermacher | ⬇ 17,672 | ♥ 5 |
| 35 | gemma-4-31B-i1-GGUF GGUFi1 | mradermacher | ⬇ 17,591 | ♥ 10 |
| 36 | Gemma-4-Novelist-Eclipse-31B-i1-GGUF GGUFi1 | mradermacher | ⬇ 17,139 | ♥ 33 |
| 37 | gemma-4-31B-it-qat-AWQ-INT4 AWQINT4QAT | cyankiwi | ⬇ 16,793 | ♥ 7 |
| 38 | Gemma-4-31B-it-DECKARD-HERETIC-Uncensored-NVFP4-SVDQuant HereticNVFP4vLLMUncensored | AEON-7 | ⬇ 15,253 | ♥ 4 |
| 39 | Gemma-4-31B-it-DECKARD-HERETIC-Uncensored-NVFP4 HereticNVFP4vLLMUncensored | AEON-7 | ⬇ 15,249 | ♥ 15 |
| 40 | Huihui-gemma-4-31B-it-qat-q4_0-unquantized-abliterated-GGUF AbliteratedGGUFQ4_0QAT | huihui-ai | ⬇ 14,903 | ♥ 24 |
Techniques in this family
GGUFHeretici1UncensoredvLLMQATAbliteratedNVFP4llama.cppARAAWQ4-bitMLXQ4_0JANGFP8
Other model families
Qwen3.8-27BQwen · 1,128 models
ERNIE-5.5-Flash-Basebaidu · 5 models
DeepSeek-V4-Prodeepseek-ai · 460 models
DeepSeek-V4.1-Flash-1017deepseek-ai · 529 models
gemma-4-12Bgoogle · 1,135 models
gemma-4-26B-A4Bgoogle · 1,131 models
gemma-4-E2Bgoogle · 1,229 models
gemma-4-E4Bgoogle · 1,189 models
LFM3-1.7BLiquidAI · 364 models
MiniMax-M2.5MiniMaxAI · 170 models
Kimi-K3moonshotai · 504 models
MiniCPM5-2Bopenbmb · 440 models
Qwen3.5-4B-BaseQwen · 555 models
Qwen3.5-9B-BaseQwen · 960 models
Qwen3.8-Flash-NextQwen · 1,230 models
Hy3tencent · 165 models
Inkling-2-Smallthinkingmachines · 35 models
Inkling-Smallthinkingmachines · 183 models
Inklingthinkingmachines · 341 models
GLM-5.3-Flashzai-org · 645 models
GLM-5.3zai-org · 1,008 models